Molluscicidal sesquiterpene lactones from species of the tribe Vernonieae (Compositae)

Article abstract of DOI:10.1002/cbdv.200800156

Schistosomiasis is caused by parasitic flatworms of the genus Schistosoma, and some snails, particularly of the genus Biomphalaria (Planorbidae), are directly implicated in the transmission of the disease. Continuing with our investigations of bioactive p

Full text of DOI:10.1002/cbdv.200800156

CHEMISTRY & BIODIVERSITY – Vol. 6 (2009)
513
Molluscicidal Sesquiterpene Lactones from Species of the Tribe Vernonieae
(Compositae)
by Susana Borkoskya), Susana Ponce de Leo´na), Gabriela Jua´reza), Manuel Gonza´lez Sierrab), and
Alicia Bardo´n*a)
a
´
) Instituto de Quꢀmica Organica, Facultad de Bioquꢀmica, Quꢀmica y Farmacia, Universidad Nacional
´
´
de Tucuman, Ayacucho 471, Tucuman 4000, Argentina
(phone/fax: þ54-381-4248169; e-mail: alisan@unt.edu.ar)
b
´
) Facultad de Ciencias Bioquꢀmicas y Farmaceuticas, Universidad Nacional de Rosario, Suipacha 531,
´
Rosario 2000, Santa Fe, Argentina
Schistosomiasis is caused by parasitic flatworms of the genus Schistosoma, and some snails,
particularly of the genus Biomphalaria (Planorbidae), are directly implicated in the transmission of the
disease. Continuing with our investigations of bioactive plant constituents, we evaluated and report in the
present article, the molluscicidal effects of 16 sesquiterpene lactones, as well as the commercial reagents
tetrahydrofuran, furfural, and furfuryl alcohol, on an adult population of B. peregrina. The natural
sesquiterpene lactones tested are characteristic constituents of species of the tribe Vernonieae, family
Asteraceae. Compounds 1–3 and 7 came from a Bolivian collection of Vernonanthura pinguis,
compounds 4 and 5 from an Argentine collection of Cyrtocymura cincta var. cincta, 6 was obtained from a
Bolivian collection of Eirmocephala megaphylla, 8–14 from an Argentine collection of Centratherum
punctatum ssp. punctatum, and compounds 15 and 16 were obtained by chemical derivatization from 5
and 14, respectively. Ten of the sesquiterpene lactones displayed moderate molluscicidal activity (LD50
100 mg/ml). Commercial reagents were inactive.
<
Introduction. – Schistosomiasis, commonly known as bilharzia, is endemic to ca. 75
countries throughout South America, Africa, and the Far East. About 250 million
people are annually infected. It is caused by parasitic flatworms of the genus
Schistosoma [1], and some snails, particularly of the genus Biomphalaria, are directly
implicated in the transmission of the disease. Incidence of schistosomiasis is increasing
as a result of the construction of dams and the introduction of irrigating schemes, which
inadvertently provide ideal breeding sites for the snail vectors. Chemotherapy is one
way of controlling this disease, though a disadvantage of the method is the high cost of
the drugs and the possibility of re-infection. The intermediate host (the mollusc)
constitutes the weakest link in the cycle of transmission and thus is the logical point of
attack to control the disease with molluscicidal agents interrupting the parasiteꢁs life
cycle and preventing infection of people in contact with water in high-risk areas. There
is an urgent need for more selective and efficient molluscicides for the control of the
snail vector, especially because the presently available compounds or formulations tend
to be general biocides, affecting many of the plants or animals (or both) in the snail
environment [2]. The World Health Organization (WHO) has given plant mollusci-
cides a high priority as tools for the integrated control of schistosomiasis due to their
low cost and rapid biodegradability [3].
ꢂ 2009 Verlag Helvetica Chimica Acta AG, Zꢃrich

514
CHEMISTRY & BIODIVERSITY – Vol. 6 (2009)
During the course of our search for biologically active compounds from plants, we
noticed that many species belonging to the tribe Vernonieae (Asteraceae) have been
used in traditional medicine [4], and have been investigated in relation to their
biological activity. They have shown to possess insecticidal [5], analgesic, anti-
ulcerogenic [6], antiparasitic [7][8], antibacterial [9], antifungal [10], cytotoxic [11],
and molluscicidal actions [12][13]. Continuing with our investigations of bioactive
plant constituents, we evaluated and report, in the present article, the toxicity of the
sesquiterpene lactones 1–14 isolated previously in our laboratory from species of the
tribe Vernonieae [14–16] on an adult population of the mentioned snail. To find the
structural requirements for activity, the synthetic derivatives 15 and 16 as well as the
reagents 17–19 were tested for their molluscicidal effects under the same experimental
conditions.
Results and Discussion. – Compounds. The sesquiterpene lactones evaluated for
their molluscicidal action are 14 highly oxygenated germacranolides characteristic of
the tribe Vernonieae. In addition, two synthetic derivatives and three commercial
reagents were also evaluated. Compounds 1–16 belong to four structural types:
hirsutinolides, 1–4 and 15, glaucolides, 5–7, goyazensolides, 8–10 and 16, and
isogoyazensolides, 11–14. Hirsutinolides and glaucolides (Figs. 1 and 2) are widespread
in the subtribe Vernoniinae of the tribe Vernonieae, and their molecules carry a
C(7)¼C(11) bond (endocyclic), a C(13)ꢀOAc group, and an ester chain at C(8).
Goyazensolides and isogoyazensolides (Figs. 3 and 4), common metabolites in species
of the subtribes Centratherinae and Lychnophorinae of the tribe Vernonieae, possess
furanone and exomethylidene-g-lactone rings in their structures [17]. Compound 15
(Fig. 5), obtained by chemical derivatization of 5 in our laboratory, was previously
isolated as a natural product [15]. The new compound 16 (Fig. 5) was obtained by
reduction of the isogoyazensolide 14 with NaBH4, and its structure was determined by
spectroscopic techniques. The molecular weight of compound 16 was established by
HR-CI-MS, which showed a quasimolecular ion peak at m/z 379.1759 ([MþH]þ )
consistent with the molecular formula C20H26O7. The presence of a OH group, an a,b-
unsaturated ketone, and two esters was deduced from the FT-IR bands at 3400, 1690,
1735, and 1760 cmꢀ1, respectively. The molecular formula accounted for eight degrees
of unsaturation, two of them were HC¼C moieties that were inferred from the typical
1H-NMR signals at d(H) 5.72 and 6.18, assigned to HꢀC(2) and HꢀC(3’), the vinyl H-
atom of the angeloyloxy (¼2-methylbut-2-enoyloxy) residue at C(8), respectively.
Fig. 1. Chemical structures of natural hirsutinolides 1–4

CHEMISTRY & BIODIVERSITY – Vol. 6 (2009)
515
Fig. 2. Chemical structures of natural glaucolides 5–7
Fig. 3. Chemical structures of natural goyazensolides 8–10
Fig. 4. Chemical structures of natural isogoyazensolides 11–14
Fig. 5. Synthetic derivatives 15 and 16

516
CHEMISTRY & BIODIVERSITY – Vol. 6 (2009)
Because three additional degrees of unsaturation are related to the presence of three
C¼O groups (13C-NMR signals at 204.4, 166.7, and 177.8), the compound must be
tricyclic. The 13C-NMR and HSQC (1H,13C correlation) spectra showed the presence of
five Me groups (two more than compound 14, the synthetic precursor), one CH2 group,
eight CH groups, and six quaternary C-atoms. Additional evidence for the presence of
an angeloyloxy chain was provided by the CI-MS spectrum in which the base peak at
m/z 279 was assigned to the loss of angelic acid (¼2-methylbut-2-enoic acid) from the
([MþH]þ ) ion (for full spectral data, see Exper. Part). The NOESY spectrum
provided information on the relative configuration, as depicted in Fig. 5. The b-
orientation of Me(15), HꢀC(5), and HꢀC(6) was evident from the correlations
observed in the NOESY spectrum (Fig. 6) among the signals of the mentioned H-
atoms. NOESY Correlations also showed that HꢀC(8) and HꢀC(11) are located on
the same side of the molecule (b), therefore, Me(13) should be a-oriented as shown in
Fig. 6.
Fig. 6. Partial NOEs observed for compound 16
Molluscicidal Effects. Sesquiterpene lactones from Vernonieae were previously
reported as natural molluscicides against collections of the Brazilian snail B. glabrata,
whereas we selected, for the evaluation of the molluscicidal activity of our
sesquiterpene lactones, the snail B. peregrina, widespread in the north of Argentina.
The bioactivity of these compounds have many times been related to the presence of an
exomethylidene-g-lactone moiety which may attack SH groups of proteins in Michael
addition reactions [18]. Among the compounds reported herein, the natural
goyazensolides 8–10 and the isogoyazensolides 11–14 possess an exomethylidene-g-
lactone group, while the remaining lactones 1–7, 15, and 16 lack this structural feature.
A quick glance at the Table that contains the results of molluscicidal activity indicates
that not all the exomethylidene-g-lactones are active; therefore, this structural feature
may be necessary but not enough for the molluscicidal activity. Our results (Table)
show that all the isogoyazensolides 11–14 tested displayed strong effects with LD50
values between 27.99 and 50.13 mg/ml, while, in hirsutinolides, glaucolides, and
goyazensolides, the activity varies depending on the functional groups attached to
the backbone of each compound. In isogoyazensolides, the exomethylidene
C(4)¼C(15) bond together with the exomethylidene-g-lactone moiety that carries a
C(7)¼C(11) bond, play an important role for the activity, as can be inferred from the
loss of the molluscicidal effect that occurred in the synthetic derivative 16 (LD50 ꢁ
100 mg/ml), in which both C¼C bonds were chemically reduced. In goyazensolides, the

CHEMISTRY & BIODIVERSITY – Vol. 6 (2009)
517
effects depended on the ester chain at C(8), the methacrylate 8 being the most active.
Glaucolides 5 and 6 with a C(1)¼O group are more active than the diepoxide 7 (Table)
that lacks that C¼O group. Interestingly, the active hirsutinolides 4 and 15 are those
which possess two vicinal OH groups on C(1) and C(10). Finally, to check the influence
of the furan moiety on the activity, we investigated the molluscicidal effects of the
furan-containing reagents 17, 18, and 19 (Fig. 7) and found that all three were inactive.
Therefore, we conclude that the activity depends on a combination of structural
features including the type of the germacranolide skeleton and the functional groups
present in the molecule.
Fig. 7. Chemical structures of commercial heterocyclic compounds 17–19
Table. Molluscicidal Activity of Compounds 1–19 against Biomphalaria peregrina
Compound
LD50 [mg/ml] (95% CL)a)
Compound
LD50 [mg/ml] (95% CL)a)
1
2
3
4
5
6
7
8
9
>100
>100
>100
11
12
13
14
15
16
17
18
19
27.99 (13.67, 37.72)
35.35 (27.93, 41.57)
50.13 (37.45, 65.80)
28.53 (10.68, 46.70)
68.48 (45.83, 99.78)
>100
>100
>100
>100
56.48 (42.21, 72.05)
38.18 (15.81, 54.75)
78.70 (44.52, 119.68)
>100
48.68 (35.22, 63.43)
88.25 (63.85, 132.84)
>100
10
a) Mortality was recorded after 24 h of exposure to the chemical. Values [mg/ml] are given as the lethal
dose 50 with its 95% confidence limit (CL).
Even when our sesquiterpene lactones are not as potent molluscicides as saponins
[2], plants of the tribe Vernonieae are widespread in the north of Argentina and are
easily available for the native population in large amounts for the preparation of
extracts containing active compounds. The present investigation represents a new step
towards the use of plant-derived molluscicides, in the hope that locally available plant
material might be applied to the control of schistosomiasis in endemic areas.
This research was supported by grants from Consejo de Investigaciones de la Universidad Nacional de
´
´
´
Tucuman (CIUNT), Agencia Nacional de Promocion Cientꢀfica y Tecnologica (ANPCyT), and Consejo
´
Nacional de Investigaciones Cientꢀficas y Tecnicas (CONICET).
Experimental Part
Plants and Compounds. Plant font, isolation, and identification of the sesquiterpene lactones
employed in the bioassays were previously reported in [14–16]. Purity of tested compounds was assessed
1
by UV, H-NMR, and MS analysis of samples purified by HPLC. Hirsutinolides are often mixed with

518
CHEMISTRY & BIODIVERSITY – Vol. 6 (2009)
glaucolides in species of the subtribe Vernoniinae. It is important to point out that only hirsutinolides can
be easily detected by UV spectroscopy due to their absorption, in MeOH/H2O solns., at lmax of 285 nm (e
10000); therefore, UV spectroscopy is an excellent technique to differentiate hirsutinolides from
glaucolides. On the other hand, it is also possible to differentiate goyazensolides from isogoyazensolides.
In the UV spectrum, the former absorb at lmax 268–269 nm and the latter at lmax 279–282 nm.
Compounds 1–3 came from a Bolivian collection of V. pinguis [14]; compound 4 from an Argentine
collection of C. cincta var. cincta [15]. Compounds 5 and 6 were present in a Bolivian collection of E.
megaphylla [15]; 7 came from a Bolivian collection of V. pinguis; 8–14 from an Argentine collection of C.
punctatum ssp punctatum [16]. The hirsutinolide 15 was obtained by chemical derivatization from the
glaucolide 5, while 16 was obtained by chemical reduction of 14 with NaBH4.
Reagents. All of them p.a. grade. Tetrahydrofuran (17) was purchased from Sintorgan, and furan-2-
carbaldehyde (18) and furan-2-methanol (19) were from Sigma-Aldrich.
Synthesis of 15 from 5. A soln. of 5 (160 mg) in CHCl3 (40 ml) was shaken in the presence of silica gel
(SiO2; 4.8 g) during 48 h at r.t. [19]. When no glaucolide A (5) was detected by TLC in the reaction
mixture, the soln. was filtered to eliminate SiO2, and the solvent was then evaporated under vacuum. The
residue was processed by HPLC on a Beckman Ultrasphere RP-8 column (5 mm, 10ꢂ150 mm) using
MeOH/H2O 4 :3, with a flow rate of 2 ml/min to yield 34.8 mg of 15 (tR 32 min). Identification of 15 was
accomplished by comparison of its spectral data with those of an authentic sample from natural origin
[15].
Synthesis of 16 from 14. To a soln. of 14 (34.2 mg, 0.1 mmol) and CoCl2 ·6 H2O (37 mg, 0.15 mmol) in
CH2Cl2/MeOH 1:1 (2 ml) at ꢀ208, NaBH4 (4.2 mg, 0.1 mmol) was added, and the mixture was stirred for
6 h [20]. The mixture was then allowed to warm to r.t., the reaction was quenched by addition of brine
(2.5 ml) and extracted twice with CH2Cl2 (2 ml each time). The org. layers were combined and dried
(Na2SO4), and the solvent was removed in vacuo. The residue was purified by HPLC using a Beckman
Ultrasphere RP-18 column (5 mm, 10ꢂ150 mm) and MeOH/H2O 4 :3 with a flow rate of 2 ml/min to yield
11.4 mg of 16, identified through its spectral features.
3,10-Epoxy-5a-hydroxy-8a-[(2-methylbut-2-enoyl)oxy]-1-oxo-germacra-2-ene-11,6-lactone (16).
Light yellow oil. [a]2D7 ¼ þ101.08 (c¼0.0037, CHCl3). IR (film): 3400 (OH), 1760 (C¼O), 1735 (ester
1
C¼O), 1690 (C¼C). H-NMR (500 MHz, CDCl3; d in ppm, J in Hz): 6.18 (qq, J¼7.5, 1.5, HꢀC(3’));
5.72 (d, J¼1.5, HꢀC(2)); 4.76 (ddd, J¼10.5, 3, 1.5, HꢀC(8)); 4.54 (d, J¼7, H ꢀC(6)); 4.19 (dd, J¼7, 1.5,
HꢀC(5)); 3.14 (qt, J¼7.5, 1.5, HꢀC(4)); 3.00 (ddd, J¼10.5, 7.5, 3, HꢀC(7)); 2.84 (d, J¼7, OH); 2.41
(dq, J¼14, 10.5, HꢀC(11)); 2.34 (dd, J¼13.5, 10.5, HaꢀC(9)); 2.17 (dd, J¼13.5, 1.5, HbꢀC(9)); 1.98 (dq,
J¼7, 1.5, MeꢀC(2’)); 1.85 (quint, J¼1.5, 3 HꢀC(4’)); 1.51 (s, Me(14)); 1.44 (d, J¼7.5, Me(15)); 1.35 (d,
J¼7, Me(13)). 13C-NMR (125 MHz, CDCl3; d in ppm): 204.4 (C(1)); 192.6 (C(3)); 177.8 (C(1’)); 166.7
(C(12)); 141.6 (C(3’)); 126.2 (C(2’)); 104.9 (C(2)); 89.6 (C(10)); 82.7 (C(6)); 77.0 (C(5)); 68.7 (C(8));
48.3 (C(9)); 46.0 (C(7)); 39.7 (C(4)); 37.9 (C(11)); 20.7 (Me(14)); 20.4 (Me(4’)); 16.2 (Me(13)); 15.9
(MeꢀC(2’)); 14.9 (Me(15)). HR-CI-MS: 379.1759 ([MþH]þ , C20H27O7þ ; calc. 379.1757).
Molluscs. The fresh water snail B. peregrina employed for the molluscicidal assay was taken from our
laboratory stock culture maintained for 4 years. The snails were reared in aquaria with dist. H2O at 26ꢃ
28, pH 7.2, under laboratory lighting conditions with normal diurnal light changes and fed on fresh leaves
of Lactuca sativa L. For water mineralization, 0.005 g/l of Ca3(PO4)2 were monthly added to the aquaria.
Molluscicidal Activity. Bioassay was assessed against B. peregrina adults according to the method
specified by WHO [21] and modified by the authors. The snails, uniform in age and size (average
diameter of the shell, 7 mm), were maintained for 24 h without feeding before the experiment.
Compounds were dissolved in MeOH and diluted with dist. H2O to reach conc. of 100, 50, and 25 mg/ml
(H2O/MeOH 98 :2). Solns. of each compound (20 ml) were then poured into 100-ml flasks, and five snails
were then placed in each flask. Control experiments were performed placing five snails in a mixture of
dist. H2O/MeOH 98 :2. After 24 h, snails were removed from the flasks, and the heart beat was observed
in a stereoscopic microscope. The mortality was then recorded. To confirm mortality of the snails, they
are placed in a beaker containing dist. H2O alone. After 24 h, their condition is re-checked. A 10 mg/ml
H2O soln. of CuSO4 was used as positive control, since it produced a 100% mortality of the snail
population. Experiments were conducted in duplicate. Data were analyzed with the Finney computer
program to determine the LD50 value at a 95% confidence interval [22].

CHEMISTRY & BIODIVERSITY – Vol. 6 (2009)
519
Supplementary Information Available. CI-MS, and 1H- and 13C-, 1H,1H-COSY, HSQC, and NOESY
NMR spectra of compound 16 are available from the corresponding author.
REFERENCES
[1] ꢄTropical Diseases: Progress in International Research 1987–1988, Ninth Program Report UNDP/
World Bank/WHO Special Program for Research and Training in Tropical Diseases (TDR)ꢁ,
Geneva, Switzerland, 1989, p. 6.
[2] A. Marston, K. Hostettmann, in ꢄMethods in Plant Biochemistryꢁ, Eds. P. M. Dey, J. B. Harborne,
Academic Press, London, 1991, p. 153.
[3] H. Kloos, F. S. McCullough, in ꢄPlant Molluscicidesꢁ, Ed. K. E. Mott, John Wiley & Sons Ltd., New
York, 1987, p. 45.
[4] P. Rasoanaivo, A. Petitjean, J. Y. Conan, Fitoterapia 1993, 64, 114.
´
´
´
[5] A. Bardon, S. Popich, D. Alvarez Valdes, C. A. N. Catalan, J. Econ. Entomol. 1999, 92, 1369.
˜
[6] V. S. Frutuoso, M. R. R. Gurjao, R. S. B. Cordeiro, M. A. Martins, Planta Med. 1994, 60, 21.
[7] T. M. Alves, T. J. Nagem, L. H. de Carvalho, A. U. Krettli, C. L. Zani, Planta Med. 1997, 63, 554.
[8] H. Fuchino, T. Koide, M. Takahashi, S. Sekita, M. Satake, Planta Med. 2001, 67, 647.
[9] G. Roos, H. Prawat, C. U. Walter, I. Klaiber, B. Vogler, J. H. Guse, W. Kraus, Planta Med. 1998, 64,
673.
[10] D. E. Wedge, J. C. G. Galindo, F. A. Macꢀas, Phytochemistry 2000, 53, 747.
[11] P. A. dos Santos, M. F. Castro Amarante, A. M. Soares Pereira, B. Bertoni, S. Castro Franca, C.
Pessoa, M. O. De Moraes, L. V. Costa-Lotufo, M. R. Pinho Pereira, N. P. Lopes, Chem. Pharm. Bull.
2004, 52, 1433.
´
[12] M. C. B. V. Alarcon, J. L. Callegari Lopes, W. Herz, Planta Med. 1990, 56, 271.
[13] D. A. Dꢀas Barros, J. L. Callegari Lopes, W. Vichnewski, J. N. Callegari Lopes, P. Kulanthaivel, W.
Herz, Planta Med. 1985, 51, 38.
´
´
[14] S. Borkosky, A. Bardon, C. A. N. Catalan, J. Dꢀaz, W. Herz, Phytochemistry 1997, 44, 465.
´
´
´
[15] S. Borkosky, D. Alvarez Valdez, A. Bardon, J. Dꢀaz, W. Herz, Phytochemistry 1996, 42, 1637.
´
´
´
[16] D. Alvarez Valdes, A. Bardon, C. A. N. Catalan, T. E. Gedris, W. Herz, Biochem. Syst. Ecol. 1998,
26, 805.
[17] W. Herz, in ꢄCompositae: Systematics. Proceedings of the International Compositae Conference,
Kew, 1994ꢁ, Eds. D. J. N. Hind, H. J. Beentje, Kew Scientist, Kew, 1996, Vol. 1, p. 229.
[18] S. M. Kupchan, M. A. Eakin, A. M. Thomas, J. Med. Chem. 1971, 14, 147.
´
´
[19] M. Martꢀnez-Vazquez, S. Sepulveda, M. A. Belmont, M. Rubio, P. Joseph-Nathan, J. Nat. Prod.
1992, 55, 884.
[20] S. O. Giordano, M. J. Pestchanker, E. Guerreiro, J. R. Saad, R. D. Enriz, A. M. Rodrꢀguez, E. A.
´
´
Jauregui, J. Guzman, A. O. M. Marꢀa, G. H. Mendel, J. Med. Chem. 1992, 35, 2452.
[21] World Health Organization. Monog. Ser. 50, 1965.
[22] D. J. Finney, ꢄProbit Analysis: Statistical Treatment of the Sigmoid Response Curveꢁ, Cambridge
University Press, London, 1964.
Received April 23, 2008